Sign inSign up

jbothma/mfmacrawl

By jbothma

•Updated over 5 years ago

Image
0

163

jbothma/mfmacrawl repository overview

⁠Scrape mfma.treasury.gov.za and build mfmamirror.github.io jekyl source files

⁠Spider Arguments

  • scrape_menu - optional - whether the menu should be scraped for data items and the links crawled futher
  • start_url - optional - a single replacement for the default start URL of the site root. Default true

⁠Project Settings

  • S3_BUCKET_NAME
  • AWS_KEY_ID
  • AWS_KEY_SECRET
  • ITEM_PIPELINES: {"mfma.pipelines.DepagingPipeline": 100,"mfma.pipelines.InternetArchiveFileArchivePipeline": 100}

⁠Set up dev environment

poetry install

Each time you run python code, you need to enter the python virtual environment, e.g.

poetry shell

⁠Run tests

poetry run pytest

⁠run scraper locally

poetry run scrapy crawl mfma -o mfma.json

Scrape a specific URL at a specific depth for debugging or something

poetry run scrapy crawl mfma -s DEPTH_LIMIT=1 -a scrape_menu=false -a start_url=http://mfma.treasury.gov.za/Circulars/Pages/default.aspx

To run without pipelines:

poetry run scrapy shell -s 'ITEM_PIPELINES={}'

⁠Run monthly

Install into cron

44 20 1 * * cd /home/pi/mfmamirror/mfmacrawl/ && trickle -d 200 -u 200 /home/pi/mfmamirror/mfmacrawl/env/bin/scrapy crawl mfma -t jsonlines -s S3_BUCKET_NAME=mfmamirror -s AWS_KEY_ID=... -s AWS_KEY_SECRET=... -s INTERNET_ARCHIVE_KEY_ID=... -s INTERNET_ARCHIVE_KEY_SECRET=... --loglevel=INFO --logfile=/home/pi/mfmamirror/mfma.log-$(date +\%Y-\%m-\%d) -o /home/pi/mfmamirror/mfma.jsonlines-$(date +\%Y-\%m-\%d) 2>&1 >> /home/pi/mfmamirror/cron.log

⁠running within docker

docker build -t mfmacrawl .
docker run --rm --env-file .env -v /var/log/mfmacrawl:/var/log/mfmacrawl -v /var/lib/mfmacrawl:/var/lib/mfmacrawl -v /var/lib/project-data-dir:/app/.scrapy mfmacrawl

Tag summary

Content type

Image

Digest

Size

157.9 MB

Last updated

over 5 years ago

docker pull jbothma/mfmacrawl